| { | |
| "best_metric": 0.9428571428571428, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148", | |
| "epoch": 7.101351351351352, | |
| "eval_steps": 500, | |
| "global_step": 148, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06756756756756757, | |
| "grad_norm": 5.423322677612305, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.3498, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12837837837837837, | |
| "eval_accuracy": 0.21428571428571427, | |
| "eval_loss": 2.089395046234131, | |
| "eval_runtime": 4.4737, | |
| "eval_samples_per_second": 15.647, | |
| "eval_steps_per_second": 1.118, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.0067567567567568, | |
| "grad_norm": 6.005563259124756, | |
| "learning_rate": 4.81203007518797e-05, | |
| "loss": 2.234, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0743243243243243, | |
| "grad_norm": 5.659995079040527, | |
| "learning_rate": 4.43609022556391e-05, | |
| "loss": 1.9304, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1283783783783783, | |
| "eval_accuracy": 0.4857142857142857, | |
| "eval_loss": 1.5130811929702759, | |
| "eval_runtime": 4.3325, | |
| "eval_samples_per_second": 16.157, | |
| "eval_steps_per_second": 1.154, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.0135135135135136, | |
| "grad_norm": 6.728376865386963, | |
| "learning_rate": 4.0601503759398494e-05, | |
| "loss": 1.4823, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.081081081081081, | |
| "grad_norm": 7.871072292327881, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 1.099, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.1283783783783785, | |
| "eval_accuracy": 0.6857142857142857, | |
| "eval_loss": 0.8322275280952454, | |
| "eval_runtime": 4.3072, | |
| "eval_samples_per_second": 16.252, | |
| "eval_steps_per_second": 1.161, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 3.02027027027027, | |
| "grad_norm": 5.636988162994385, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 0.7533, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0878378378378377, | |
| "grad_norm": 4.6487226486206055, | |
| "learning_rate": 2.9323308270676693e-05, | |
| "loss": 0.4725, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.1283783783783785, | |
| "eval_accuracy": 0.7714285714285715, | |
| "eval_loss": 0.5527699589729309, | |
| "eval_runtime": 4.2206, | |
| "eval_samples_per_second": 16.585, | |
| "eval_steps_per_second": 1.185, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.027027027027027, | |
| "grad_norm": 2.4070096015930176, | |
| "learning_rate": 2.556390977443609e-05, | |
| "loss": 0.4228, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.094594594594595, | |
| "grad_norm": 7.774806499481201, | |
| "learning_rate": 2.1804511278195487e-05, | |
| "loss": 0.3629, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.128378378378378, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.3821955621242523, | |
| "eval_runtime": 4.4076, | |
| "eval_samples_per_second": 15.882, | |
| "eval_steps_per_second": 1.134, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 5.033783783783784, | |
| "grad_norm": 6.296115398406982, | |
| "learning_rate": 1.8045112781954888e-05, | |
| "loss": 0.248, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.101351351351352, | |
| "grad_norm": 7.289224147796631, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.196, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.128378378378378, | |
| "eval_accuracy": 0.9142857142857143, | |
| "eval_loss": 0.30108556151390076, | |
| "eval_runtime": 3.9981, | |
| "eval_samples_per_second": 17.508, | |
| "eval_steps_per_second": 1.251, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 6.04054054054054, | |
| "grad_norm": 3.051034450531006, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.1786, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.108108108108108, | |
| "grad_norm": 4.523892879486084, | |
| "learning_rate": 6.766917293233083e-06, | |
| "loss": 0.1357, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.128378378378378, | |
| "eval_accuracy": 0.9142857142857143, | |
| "eval_loss": 0.2538822889328003, | |
| "eval_runtime": 4.2283, | |
| "eval_samples_per_second": 16.555, | |
| "eval_steps_per_second": 1.183, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 7.047297297297297, | |
| "grad_norm": 0.9887262582778931, | |
| "learning_rate": 3.007518796992481e-06, | |
| "loss": 0.0847, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.101351351351352, | |
| "eval_accuracy": 0.9428571428571428, | |
| "eval_loss": 0.19233502447605133, | |
| "eval_runtime": 4.6688, | |
| "eval_samples_per_second": 14.993, | |
| "eval_steps_per_second": 1.071, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 7.101351351351352, | |
| "step": 148, | |
| "total_flos": 2.915999166844109e+18, | |
| "train_loss": 0.8118188864475971, | |
| "train_runtime": 290.0311, | |
| "train_samples_per_second": 8.165, | |
| "train_steps_per_second": 0.51 | |
| }, | |
| { | |
| "epoch": 7.101351351351352, | |
| "eval_accuracy": 0.8451612903225807, | |
| "eval_loss": 0.44350922107696533, | |
| "eval_runtime": 9.9685, | |
| "eval_samples_per_second": 15.549, | |
| "eval_steps_per_second": 1.003, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 7.101351351351352, | |
| "eval_accuracy": 0.8451612903225807, | |
| "eval_loss": 0.44481009244918823, | |
| "eval_runtime": 9.6598, | |
| "eval_samples_per_second": 16.046, | |
| "eval_steps_per_second": 1.035, | |
| "step": 148 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 148, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.915999166844109e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |